final compiled slides - wef

2/14/2017

1

2/14/2017

2

An MRRDC Short Course Biofilms: Principles and

Advanced Model-Based DesignFebruary 15th, 2017

1:00 – 3:00 PM, Eastern Standard Time

How to Participate Today

• Audio Modes

• Listen using Mic & Speakers

• Or, select “Use Telephone” and dial the conference (please remember long distance phone charges apply).

• Submit your questions using the Questions pane.

• A recording will be availablefor replay shortly after thiswebcast.

2/14/2017

3

Today’s Moderator

John B. Copp Ph.D.Primodal Inc.Hamilton, Ontario

Biofilms – Feb. 15, 2017

• Topics:

• Introduction to Critical Biofilm Concepts• Capturing Biofilm Concepts w Modelling • Biofilm Modelling Case Studies

An MRRDC Short CourseBiofilms: Principles and Advanced

Model-Based Design

2/14/2017

4


• Speakers:

Leon Oliver TanushDowning Schraa Wadhawan

CH2M InCTRL Solutions Dynamita


Model-Based Design

Next Speaker

Leon Downing, PhD, PESenior TechnologistCH2MMilwaukee, Wisconsin, USA

2/14/2017

5

Biofilm Reactors:What’s the BIG Deal?

Leon Downing, PhD, PECH2M

Some of the first technologies for wastewater management relied on ‘slime’

1890sTrickle sewage over rocksGrow slimeCleans water

2/14/2017

6

So is it slime? Or fixed film? Or Biofilm?• Biofilm is the official terminology for WEF and the

larger scientific community

• “cells immobilized at a substratum and frequently embedded in an organic polymer matrix [EPS] of microbial origin” - Characklis and Marshall 1990

Why does this warrant an entire webcast?

Source: Center for Biofilm Engineering at MSU-Bozeman

2/14/2017

7

Key differences from suspended growth (e.g. activated sludge)

• Mass transfer limitation

• Zone specific ecology

• Settleability of solids

• Biomass control and quantification

How does stuff get into a biofilm?

2/14/2017

8

How does stuff get into a biofilm?Smaller mass transfer boundary layer (MTBL): more transfer into biofilm


Higher concentration in bulk liquid:more transfer into biofilm

Smaller mass transfer boundary layer (MTBL): more transfer into biofilm

2/14/2017

9


Higher concentration in bulk liquid:more transfer into biofilm

Smaller mass transfer boundary layer (MTBL): more transfer into biofilm

Thickness is important for biofilm performance and health

Despite mass transfer limitation, same “bugs” are present

2/14/2017

10

Mass transfer is the most critical consideration for biofilms

• Also important for activated sludge systems!

Mass transfer is the most critical consideration for biofilms

• Also important for activated sludge systems!

Anoxic centerAnoxic center

Mass transfer limitations in floc allows for the development of anoxic conditions in aerobic tanks, resulting in simultaneous nitrification and denitrification

2/14/2017

11

But how do the solids settle?

• Doesn’t settle poorly, they are just small solids

Controlling the biofilm is a key operational consideration• Controlling rates Decrease MTBL thickness Increase bulk liquid

concentrations

• Controlling biofilm thickness Prevent excess thickness and

excess sloughing Not too thick, not too thin,

but just right

2/14/2017

12

Given our key differences, what are the key design/modeling parameters?

• Biofilm thickness

• Mass transfer boundary layer thickness

• Biofilm growth and detachment

• Biofilm stratification and layers

Given our key differences, what are the key design/modeling parameters?

• Biofilm thickness

• Mass transfer boundary layer thickness

• Biofilm growth and detachment

• Biofilm stratification and layersMixing is a critical design parameters

2/14/2017

13

Mixing is a critical part of biofilm reactor design and modeling

Flushing water/arm speed

Aeration and mechanical mixing

Aeration and mechanical mixing

What are our key design parameters for biofilms?• Activate sludge

SRT limited system Hydraulic retention

time (HRT) Sludge retention time

(SRT) Volumetric loading

rate DO setpoints of less

than 2 mg/L

• Biofilm reactors Mass transfer limited

system Hydraulic retention time

(HRT) Biofilm thickness MTBL assumptions Attachment surface area Media specific surface

area Area loading rate DO setpoints of 4 to 5

mg/L

2/14/2017

14

Next Speaker

Oliver Schraa, M.Eng.Chief Technical OfficerinCTRL Solutions Inc.Oakville, Ontario, Canada

Modeling Biofilm ReactorsOliver Schraa

inCTRL Solutions Inc., Oakville, Ontario

2/14/2017

15

Outline

• Introduction to activated sludge modeling

• Additional processes modeled in biofilm reactors

• Guidance on model setup and calibration

• Example

• Summary

Activated Sludge ModelingBasic Structure of a Typical Activated

Sludge Model

2/14/2017

16

Basic reactor mass balance:

Accumulation = Transport + Generation



Generation or Consumption

2/14/2017

17



,, ,

Assumptions: Completely-mixed, constant volume, influent flow = effluent flow

Substrate utilization by biomass



,, ,

, ,

,

Monod kinetics

2/14/2017

18

Additional Processes Modeled in Biofilm Reactors

• Diffusion of soluble components into and within biofilm

• Varying biofilm thickness due to growth and decay, attachment of particles, and detachment of particles

• Stratification of biofilm into regions with different organisms and redox conditions

Conceptual biofilm structure for model development

2/14/2017

19

Conceptual biofilm structure for model development

Assumptions:• Biofilm consists of layers

(allows modeling of concentration gradients)

• Each layer is a flat uniform sheet

• Diffusion is only method of soluble component transport within biofilm

• Diffusion is in direction perpendicular to biofilm

Biofilm reactor mass balances: soluble components• Bulk phase:


2/14/2017

20



DiffusionAdvection

,, , ,

Substrate utilization by biomass


Diffusion across liquid boundary layer


,, , ,

2/14/2017

21

Diffusion Across the Boundary Layer

Figure from Modelling Biofilms, Morgenroth (2008)

Diffusion across the liquid boundary layer (quiescent zone) is rate limiting for

substrate conversion

Bulk Liquid

Biofilm

Boundary Layer



,

Based on Fick’s First Law

2/14/2017

22



,

Diffusion rate is proportional to concentration difference between bulk liquid and at

the biofilm surface



,

Diffusion rate is inversely proportional to the liquid boundary layer thickness

2/14/2017

23

Biofilm reactor mass balances: soluble components• Within biofilm:



Fick’s first law of diffusion used to describe diffusion

,,


Diffusion that varies with biofilm depth

2/14/2017

24



Becomes a partial differential equation

,,

,,



Diffusion term discretized in space by splitting biofilm into layers

2/14/2017

25

Biofilm reactor mass balances: particulate components• Within biofilm:


Diffusion

Attachment anddetachment

Bacterial growth and decay

Biofilm reactor mass balances: particulate components• Within biofilm:


Becomes a partial differential equation

, ,,

From Wanner et al. (2006)

2/14/2017

26

Balance on biofilm thicknessChange in thickness = Velocity of expansion or contraction due to

biomass growth or decay + Attachment velocity –Detachment velocity

, ,

From Wanner et al. (2006)

Model Solution

• Solve system of partial differential equations (discretized into ordinary differential equations)

• Solve model using numerical integration (dynamic) or nonlinear algebraic solver (steady-state)

• Difficult system to solve numerically because of competing slow and fast processes e.g. biomass growth versus diffusion rate Specially designed solvers can help but biofilm reactor

simulations are typically very time-consuming

2/14/2017

27

Model CalibrationGeneral Steps with Any Activated

Sludge Model

The GMP Unified Protocol1. Project Definition2. Data Collection & Reconciliation3. Plant Model Set-Up4. Calibration/Validation5. Simulation & Results Interpretation

Agreement on the objectives and budget?

Requirements (data, model accuracy, …)

Problem statement

Objectives

Process flow diagram (process units, boundaries)

Data collection (physical, operational, influent, effluent)

Data and process evaluation

Calibration / validation adequate ?

Selection of the parameters and stop criteria

Plant model building (input specification, sub-model selection)

Calibration

Simulation and result interpretation

Functional evaluation

Report

Objectives achieved?

Model adequate?

Data and process validated?

Validation

Data sufficient and validated?

Calibration successful?

yes

no

yes

yes

yes

yes

yes

yes

no

no

no

no

no

no

no

Additional data collection

(direct parameter

measurements, sampling

campaign, missing data)

Sensible outputs?

yes

Validation successful?

yes

no

PR

OJE

CT

DE

FIN

ITIO

N

DA

TA

CO

LL

EC

TIO

N A

ND

EV

AL

UA

TIO

N

MO

DE

L S

ET

UP

AN

D E

VA

LU

AT

ION

CA

LIB

RA

TIO

N A

ND

VA

LID

AT

ION

SI

MU

LA

TIO

N A

ND

RE

SUL

T

INT

ER

PR

ET

AT

ION

Results sufficient? no

https://iwa-gmp-tg.irstea.fr

2/14/2017

28

• BNR Calibration Procedure Hydraulic model (tanks in series, flow splits) Influent characterization Sludge production (first COD and then TSS) Nitrification Denitrification Phosphorus removal

Iterative procedure!

Step 4: Calibration and Validation

Model Setup and CalibrationAdditional Steps with Biofilm Models

2/14/2017

29

Model Setup and Calibration

• Key Additional Steps for Biofilm Reactors Reactor physical parameters: Carrier surface area and reactor fill fraction Density of biofilm Number of biofilm layers

Calibration parameters: Liquid boundary layer thickness Biofilm detachment rate Diffusion reduction in biofilm

Model Setup

• Specify details of media

From Wallis-Lage et al. (2006)

2/14/2017

30

Model Setup

• Specific carrier surface area: Surface area per volume of media

Ranges from 50 to 4,000 m2/m3 (Morgenroth, 2008)

• Reported range for MBBR media is 400 to 1,200 m2/m3 (Weiss et al., 2005)

Data provided by manufacturers

Model Setup

• Fraction of reactor volume filled by media: Depends on technology 30 to 60% is typical for MBBRs and IFAS Also need water volume displaced by media

2/14/2017

31

Model Setup

• Specify details of biofilm

Photo from Ødegaard (1999)

• Density of biofilm: Specify water content and dry density

Biofilm is typically between 3 and 7% dry solids

Density of dry solids ranges from 10 to 100 g/L (Melo, 2005)

Model Setup

2/14/2017

32

• Number of biofilm layers Affects soluble profile through biofilm 3 to 5 layers typical in most simulators Fewer layers = faster simulations but less

accurate profiles

Lewandowski Z and Boltz JP (2011) Biofilms in Water and Wastewater Treatment. In: Peter Wilderer (ed.) Treatise onWater Science, vol. 4, pp. 529–570 Oxford: Academic Press.

Model Setup

Model Calibration

• Estimation of liquid boundary layer


2/14/2017

33

• Liquid boundary layer Affects rate of diffusion into biofilm

Diffusion across biofilm is usually the rate limiting step

Thickness directly related to hydrodynamic conditions Higher wastewater and air flowrates and more

intense mixing lead to thinner boundary layers

Model Calibration

• Liquid boundary layer Ranges between 20 and 1,500 μm (Morgenroth, 2008)

For exisiting systems: calibrate using soluble concentrations in reactors, or estimate with empirical correlations

For design: calibrate to manufacturer or pilot data

Model Calibration

2/14/2017

34

• Biofilm detachment rate Impacts biofilm thickness and SRT

Typical expression used:• Wanner and Gujer (1986)

Model Calibration

Detachment rate is proportional to biofilm density and square of

biofilm thickness

• Diffusion rates Default model values taken from chemical

engineering literature

Reduction in biofilm: Diffusion reduced in biofilm due to interactions

with particulates Ratio of diffusion in biofilm to diffusion in liquid

phase is typically between 0.3 and 0.8 Typically leave at default value

Model Calibration

2/14/2017

35

What are the most important biofilm-specific model inputs?• Physical chacteristics of media:

Carrier surface area, water displaced by media, and reactor fill fraction

Density of biofilm Number of biofilm layers

• Parameter requiring calibration: Liquid boundary layer thickness

Important concepts for design and modeling of biofilm reactors

• Mass transfer is typically the rate-limiting step

• Identity of the limiting substrate changes over length of a biofilm reactor

• Can have different redox conditions within different regions of a biofilm

• Can have stratifcation of organisms within different regions of a biofilm

2/14/2017

36

Example• Using a model to study impact of mass transfer resistance

MBBR Plant for BOD removal and nitrification

Four MBBR’s in series

Example• Using a model to study impact of mass transfer resistance

DO = 2 mg/L in all tanks


Little nitrification: heterotrophs dominate in all tanks

Significant nitrification in last two tanks

2/14/2017

37

Example


Significant nitrification in second tank: Nitrification becomes significant once oxygen penetrates far enough into biofilm and COD is low enough

Example

Adapted by Morgenroth (2008) from Henze et al. (2002)

Simulation agrees with experimental data: Nitrification rate is higher when organic substrate penetratation into biofilm is low relative to oxygen penetration

Ratio of organic substrate penetration to O2 penetration

2/14/2017

38

Summary• Biofilm reactor modeling is similar to suspended

growth modeling but more information is required for model setup

• Biofilm reactors are mass transfer limited and models are useful in studying how this impacts reactor design and control

• The key input parameters are: Carrier surface area, water displaced by media, and

reactor fill fraction Density of biofilm Number of biofilm layers Liquid boundary layer thickness

What is Next?Using Biofilm Reactor Models as Part

of Plant Design and Optimization

2/14/2017

39

Oliver SchraaM.Eng.

inCTRL Solutions Inc.Canada

Email: [email protected]: www.inCTRL.ca

Presenter Contact Information

ReferencesHenze,M., Harremoës,P., Jansen,J.l.C. and Arvin,E. (2002). Wastewater Treatment. 3rd edition. Springer, Berlin.

Lewandowski, Z. and Boltz, J.P. (2011). Biofilms in Water and Wastewater Treatment. In: Peter Wilderer (ed.) Treatise on Water Science, vol. 4, pp. 529–570 Oxford: Academic Press.

Morgenroth, E. (2008). Modelling Biofilms. In: Henze, M., van Loosdrecht, M.C.M., Ekama, G.A., and Brdjanovic (ed.) Biological Wastewater Treatment: Principles, Modelling and Design., IWA Publishing, London, UK.

Ødegaard, H. (1999). The Moving Bed Biofilm Reactor. In: Igarashi,T., Watanabe,Y., Asano,T. and Tambo, N. (ed.) Water Environmental Engineering and Reuse of Water, Hokkaido Press 1999, p. 250-305.

Wallis-Lage, C., Johnson, T., Hemken, B. and Sabherwal B. (2006). New Technologies Force Change from Traditional Design-Bid-Build Strategy. Proceedings of WEFTEC 2006, Oct. 21-25, Dallas, TX, USA.

Wanner,O., Eberl,H.J., Morgenroth,E., Noguera,D.R., Picioreanu,C., Rittmann,B.E. and van Loosdrecht,M.C.M. (2006). Mathematical Modeling of Biofilms, IWA Publishing, London, UK. Series: Scientific and Technical Report Series Report No. 18.

2/14/2017

40

Next Speaker

Tanush Wadhawan, Ph.D.Dynamita,Toronto, Ontario, Canada

Modeling applications for biofilm design and operations

Tanush Wadhawan, PhDDynamita

2/14/2017

41

Key outline points

• Introduction to model applications

• Basic model/design requirements

• MBBR case study/applications

• Summary

Modeling applications and guidelines

• Design

• Upgrade

• Optimization

Modified from https://iwa-gmp-tg.irstea.f

2/14/2017

42

Identifying design/upgrade requirements

SITE-SPECIFIC

WASTEWATER CHARACTERIST

TEMPERATUREEFFLUENT SAMPLINGPROCEDURES

EFFLUENT CRITERIA

LOCAL ENERGY COSTS

AVAILABLE LAND

RETROFIT, UPGRADE, OR NEW DESIG

NO ONE SIZE FITS ALL

Wastewater characterization

Total Kjeldahl Nitrogen

Total phosphorus

Total chemical oxygen demand

Soluble Colloidal Particulate

•Soluble biodegradable organics• N removal performance• Anoxic tank size• Aeration taper• P removal performance

•Particulates• Sludge production• MLSS• Clarifier sizingBiodegradable and unbiodegradable portion

Ammonia + Organic-N

Ortho-P + Organic P

TSS, VSS, BOD, pH, alkalinity…..

2/14/2017

43

• Pre-screening of inert material

• Length-to-width ratio of 0.5:1 - 1.5:1

• Carrier fill - 25 to 67%

Basic design requirements

• Coarse bubble aeration system

• Sieve to retain media

• Mechanical mixing

• Solids separation

http://www.worldwaterworks.com

6 screensTank

Aeration gridCoarse bubble diffusers

Inlet

Steady state simulations• Long term averaged performance Average annual and average monthly Average weekly – Caution! (facility might not be

at steady state)

• Good for initial process design and sizing

• Facilities are inherently dynamic (especially nutrient removal processes) Use peaking factors in design procedure Using dynamic modeling

2/14/2017

44

Dynamic simulations

• Anticipating peak loading conditions

• Refining the design

Aeration/blower design

Steady state8816 SCFM

DynamicPeak 9900 SCFM

Steady state

Solids loadinglbs/ft2/d

Effluent TSSgTSS/m3

Clarifier design

Dynamic simulations

• Diurnal• Weekly, monthly, and seasonal modelling• Maximum Month “Birthday Cake” analysis• Dynamic response to storm water event

TCOD gCOD/m3

TKN gN/m3

Q MGD

2/14/2017

45

Modeling MBBR applications

A. Case studies1. Broomfield WWTP, Colorado, USA2. James river treatment plant, VA, USA

B. Other applications1. Volumetric loading verses surface area

loading2. Comparing footprint of an MBBR versus and

activated sludge process3. Evaluating robustness of an MBBR process

versus and activated sludge process4. Improving capacity of an MBBR system

Broomfield WWTP, Colorado

2/14/2017

46

Broomfield wastewater treatment plant (BWWTP)

Biofilm reactors, Chapter 11, WEF MOP No. 35.

BWWTP configuration

• IFAS plant operated as A2O configuration

• Kaldnes K1 media with 30% fill

• Biofilm surface area 500 m2/m3 Effective surface area 150 m2/m3

• Total SRT 5 days, aerobic SRT 3.25 days

Total volume 1.853 MG

15% 20% 65%40% RAS

150% IR

2/14/2017

47

Primary effluent

• Daily flow and temperature measurements

• Temperature change is crucial• 19 C to 13 C during December

• BOD, TSS, NHx-N, NOx-N weekly

• COD/BOD ratio once a month (2.7)

Operational data collection• Media is dried, weighed and compared to

bare media for biofilm growth

• Biofilm thickness should be measured occasionally (at least visual inspection)

• The biofilm is thinner in the winter than in the summer

• First aerobic reactor has thicker (1 ± 0.2 mm) biofilm that second aerobic reactor (0.6 ± 0.2 mm)

2/14/2017

48

MBBR operational dataValues Units Comments

Suspended biomass, MLSS 1630 g/m3

Aerobic 1 fixed biomass 2050 g/m3

Aerobic 2 fixed biomass 1104 g/m3

Volume per cell 2271 m3 Design

Effective surface area 150 m2/m3 Manufacturer's data

Total surface area per cell 340650 m2 Calculated

Biomass Aerobic 1 per surface area 13.7 kg/1000 m2

Biomass Aerobic 2 per surface area 7.4 kg/1000 m2

Biofilm thickness Aerobic 1 1.1 mm

Biofilm thickness Aerobic 2 0.6 mm

Calculated from

density 12.5 kg/m3

Calculated

Measured data Red are model inputs

DO - 4.3 mgO2/l DO - 5.6 mgO2/l

Calibration process

• Garbage “In” = Garbage “Out” Very important to get the influent

characterization correct Temperature is quite important

- Nitrification rates (every 10 C increase rates double)- Models use Arrhenius equation

Estimating biofilm input parameters from data or manufacture’s data

• Using default setting of the models will simulate reasonably

2/14/2017

49

Nitrate/Ammonia effluent

NO3-N measured data (gN/m3)

- Model NO3-N (gN/m3)

NH4-N measured data (gN/m3)

- Model NHx-N (gN/m3)

Avg. NHx-N <0.2 gN/m3Avg. NO3-N 14.4 gN/m3

MLSS gXTSS/m3

Avg. MLSS 1673 gXTSS/m3- Model MLSS (gXTSS/m3)

MLSS measured data (gXTSS/m3)

2/14/2017

50

Identifying limitations

NOx-N

- Model NO3-N (gN/m3)• No data of NO3-N profile to verify• If measurements concur with the

model, then anoxic tanks are biomass limited

• Adding media to anoxic tank can improve capacity and performance

@ 150% IR

This is an example of a significant finding from a simulation study that can help improve design and operation of a plant.

Summary from BWWTP case study

• Influent and operational data Biofilm specific mass Biofilm thickness Specific surface area

• As long as most of the inputs are correct, models will give reliable outputs

• The modeling study helped identify design limitations and areas for operational improvement

2/14/2017

51

Lets try to upgrade!

• When a model is well calibrated to existing operation then scenario runs are quite beneficial Upgraded Anoxic2 from activated sludge to

IFAS by 30% K1 media Increased media to 60% Upgraded both anoxic cells with 30% K1

media

30% fill

60% fill

30% fill 30% fill

Default model inputs usedFor more accurate representation get information from published work or manufacturers

2/14/2017

52

Scenario runsScenario runs Average NO3‐N Unit

December 2006 simulation result 3.10

30% fill 1.15

60% fill 0.82

30% fill both cells 0.34

gN/m3

Improves TN removal

James river WWTP, VA, USA

2/14/2017

53

James river treatment plant (JRTP)

• One of nine major treatment plants operated by the Hampton Roads Sanitation District (HRSD) in southeast Virginia

JRTP configuration

• Initial design activated sludge

• Upgraded to pre-anoxic zones + IFAS (MLSS + media)

• The plastic media is retained by cylindrical screens

2/14/2017

54

Dynamic flow

Daily average flow (MGD)Treats an average daily flow of 15.9 MGDMax Month flows are 20 MGD and

Dynamic wastewater composition

Daily average TCOD (gCOD/m3)

Daily average TKN (gN/m3) Daily average TP (gP/m3)

2/14/2017

55

Dynamic temperature

Daily temperature (C)

JRTP configuration

Specific mass changes seasonallyPrevious example was only for December

2/14/2017

56

Biofilm specific mass profiles (gTSS/m2)

AE1 measured data AE2 measured data

-AE1 simulation-AE2 simulation

Less fixed mass during summer ca. 27 C

Higher fixed mass during winter 13 C

Profiles in biofilm reactors

AE2

DO in biofilm layersAE2

DO in biofilm layersAE2

Layer 3,4

Nitrate in biofilm layer

Ammonia in biofilm layer

AE2Nitrate in biofilm layer

Ammonia inbiofilm layer

2/14/2017

57

Effluent nitrogen species

- TN (gN/m3)

- SNHx(gN/m3)

- SNOx (gN/m3)

Summary from JRTP case study

• IFAS upgrade increased biomass in the tank

• Able to achieve more treatment in the same volume and meet low TN limits

• Calibrated models can help understand the limiting conditions for nitrification

2/14/2017

58

Increasing DO when limiting• DO gets limiting in the

biofilm layers• Limiting nitrification• Increased the bulk DO

concentration during the 269-309 days by 1 gO2/m3

• Increased NO3-N by 1 gN/m3

Other applicationsEvaluating biofilm applications using

simulation

2/14/2017

59

Volumetric loading verses surface area loadingSize and surface area of the media does not matter as long as surface loading rates are the same.

Ødegaard, 2000

Model evaluation

• Same influent and default MBBR parameters

250 m2/m3

500 m2/m3

750 m2/m3

2/14/2017

60

Same surface area loading rate

• 210, 420, and 630 gCOD/m3 to achieve same surface area loading

Surface area loading Volumetric loading

Surface area removal Volumetric removal

Same volumetric loading rate

• 420 gCOD/m3 to achieve same volumetric loading

Surface area loading Volumetric loading

Surface area removal Volumetric removal

2/14/2017

61

Comparing footprint

1.

2.

Mass of TSS 47,264 lbsEffluent nitrate 22 gN/m3

Temp 20 C, fully nitrifying, SRT 5 days Activated sludge plant

Volume 2.6 MG

Volume 2.1 MG

Temp 20 C, fully nitrifying, 50% fill MBBR plant

Mass of TSS 47,263 lbsEffluent nitrate 21 gN/m3

19% reduction

MBBR

Evaluating robustness of an MBBR process versus and activated sludge process

MBBR


MBBR

2/14/2017

62


1. Diurnal flow for a small plant2. Phosphorus limitation day 10-15

MBBR

Flow (MGD)

TP (g P/m3)


• Stable nitrification• Less sensitive to peaks• Not susceptible to washout

MBBR NOx-N (gN/m3)

Activated sludgeNOx-N (gN/m3)

P limitation

2/14/2017

63

Improving capacity

• Diurnal flow

• Increased nitrogen load by 16%

• Identify desired carrier fill


MBBR

Improving capacity of an MBBR system

• 0-5 days Fill percentage 50%

TN load 985-2432 lbsN/d

• 5-15 days Fill percentage 50%

TN load 1146-2828 lbsN/d

Effluent ammonia g N/m3

Effluent nitrate g N/m3

• 15-25 days Fill percentage 65% TN load 1146-2828 lbsN/d

Meeting effluent ammonia target < 1 gN/m3

2/14/2017

64

Quite a powerful tool

• Substrate profiles

• Biomass competition

• Desired bulk DO concentration

• Peak AOTR demand

• Desired fill percentage

By the power vested in me by the MODELS I pronounce you designed/upgraded/optimized!

Summary

• Existing models follow a variety of design guidelines and match experimental data including full-scale plant operation

• Garbage “In” means garbage “Out”

• Identifying process limitation

• Well calibrated models help useful scenario runs for design and operation improvements

• Other applications Smaller footprint Robust performance Increase capacity

• These applications can be used on a single plant to uogradefor nutrient removal

2/14/2017

65

Good reads!

Ødegraard, H, Gisvold, B, Strickland, J. (2000). The influence of carrier size and shape in the moving bed biofilm process. Water Science & Technology, 41(4-5), 383-342.

Thank you!Questions?

[email protected]

2/14/2017

66


• Final Q & A:

Biofilms Leon Downing CH2M

Modeling Oliver Schraa InCTRL Solutions

Application Tanush Wadhawan Dynamita


Model-Based Design

How to Participate Today

• Audio Modes

• Listen using Mic & Speakers

• Or, select “Use Telephone” and dial the conference (please remember long distance phone charges apply).

• Submit your questions using the Questions pane.

• A recording will be availablefor replay shortly after thiswebcast.

final compiled slides - wef

Documents